pcsx_rearmed.git
7 years agorelease r12 r12
notaz [Sat, 24 Dec 2011 01:26:20 +0000 (03:26 +0200)]
release r12

7 years agopsx_gpu: bugfix
Exophase [Sat, 24 Dec 2011 01:10:16 +0000 (03:10 +0200)]
psx_gpu: bugfix

7 years agopsx_gpu: fix sprite sign extend
Exophase [Fri, 23 Dec 2011 22:13:52 +0000 (00:13 +0200)]
psx_gpu: fix sprite sign extend

7 years agogte_neon: use more accurate division
notaz [Fri, 23 Dec 2011 19:48:30 +0000 (21:48 +0200)]
gte_neon: use more accurate division

texture placement issues in Chrono Cross otherwise with the new GPU

7 years agopsx_gpu: some argument checks
Exophase [Fri, 23 Dec 2011 16:51:42 +0000 (18:51 +0200)]
psx_gpu: some argument checks

7 years agopsx_gpu: fix sprite coordinates
Exophase [Fri, 23 Dec 2011 16:05:16 +0000 (18:05 +0200)]
psx_gpu: fix sprite coordinates

7 years agopsx_gpu: texture cache fix2
Exophase [Fri, 23 Dec 2011 00:43:58 +0000 (02:43 +0200)]
psx_gpu: texture cache fix2

7 years agopsx_gpu: texture cache fix, updates
Exophase [Fri, 23 Dec 2011 00:47:19 +0000 (02:47 +0200)]
psx_gpu: texture cache fix, updates

7 years agogpu_neon/psx_gpu: update texture_window_settings
notaz [Fri, 23 Dec 2011 00:46:52 +0000 (02:46 +0200)]
gpu_neon/psx_gpu: update texture_window_settings

7 years agopsx_gpu: fix divide by 0
Exophase [Fri, 23 Dec 2011 00:45:21 +0000 (02:45 +0200)]
psx_gpu: fix divide by 0

7 years agogpu_neon/psx_gpu: fix bad masking/alignment fault
notaz [Wed, 21 Dec 2011 20:37:56 +0000 (22:37 +0200)]
gpu_neon/psx_gpu: fix bad masking/alignment fault

7 years agogpu_neon: integrate Exophase's rasterizer
notaz [Tue, 20 Dec 2011 21:56:35 +0000 (23:56 +0200)]
gpu_neon: integrate Exophase's rasterizer

7 years agogpu_neon/psx_gpu: remove unused psx_gpu_alt
notaz [Tue, 20 Dec 2011 21:49:22 +0000 (23:49 +0200)]
gpu_neon/psx_gpu: remove unused psx_gpu_alt

saves a lot of address space

7 years agogpu_neon/psx_gpu: rename a define
notaz [Tue, 20 Dec 2011 21:19:52 +0000 (23:19 +0200)]
gpu_neon/psx_gpu: rename a define

7 years agogpu_neon/psx_gpu: use external vram
notaz [Tue, 20 Dec 2011 21:14:42 +0000 (23:14 +0200)]
gpu_neon/psx_gpu: use external vram

7 years agoadd NEON GPU rasterizer
Exophase [Tue, 20 Dec 2011 21:07:20 +0000 (23:07 +0200)]
add NEON GPU rasterizer

7 years agogpu_unai: add 4bpp asm sprite blitter
notaz [Tue, 20 Dec 2011 00:38:53 +0000 (02:38 +0200)]
gpu_unai: add 4bpp asm sprite blitter

7 years agogpu_neon: make use of unai's lineskip mode
notaz [Mon, 19 Dec 2011 21:32:49 +0000 (23:32 +0200)]
gpu_neon: make use of unai's lineskip mode

7 years agogpu_neon: fix test mode of unai renderer
notaz [Mon, 19 Dec 2011 21:29:42 +0000 (23:29 +0200)]
gpu_neon: fix test mode of unai renderer

7 years agoupdate release packaging
notaz [Sun, 11 Dec 2011 16:46:52 +0000 (18:46 +0200)]
update release packaging

use gpu_neon emu code for everything

7 years agofrontend: menu: show savestate date
notaz [Sun, 11 Dec 2011 00:32:08 +0000 (02:32 +0200)]
frontend: menu: show savestate date

7 years agogpu_neon: update for armv5pollux build
notaz [Sat, 10 Dec 2011 21:52:12 +0000 (23:52 +0200)]
gpu_neon: update for armv5pollux build

7 years agogpu_neon: tune frameskip further, allow >1
notaz [Sat, 10 Dec 2011 21:08:49 +0000 (23:08 +0200)]
gpu_neon: tune frameskip further, allow >1

7 years agodrc: fix bad gte unneeded reg assumption
notaz [Sat, 10 Dec 2011 17:20:33 +0000 (19:20 +0200)]
drc: fix bad gte unneeded reg assumption

7 years agofrontend: menu: limit enum text length
notaz [Sat, 10 Dec 2011 23:32:00 +0000 (01:32 +0200)]
frontend: menu: limit enum text length

7 years agofrontend: make analogs configurable
notaz [Fri, 9 Dec 2011 23:57:29 +0000 (01:57 +0200)]
frontend: make analogs configurable

7 years agoadd vibration support for Caanoo
notaz [Fri, 9 Dec 2011 16:35:09 +0000 (18:35 +0200)]
add vibration support for Caanoo

7 years agogpu_neon: fix unai rasterizer
notaz [Thu, 8 Dec 2011 22:31:22 +0000 (00:31 +0200)]
gpu_neon: fix unai rasterizer

it can't work directly on the list because modifies it

7 years agogpu_neon: propagate configs to rasterizers
notaz [Thu, 8 Dec 2011 21:09:36 +0000 (23:09 +0200)]
gpu_neon: propagate configs to rasterizers

7 years agocdrom: don't emulate seek when set to same loc
notaz [Wed, 7 Dec 2011 23:26:37 +0000 (01:26 +0200)]
cdrom: don't emulate seek when set to same loc

fixes C-12 (needs at least 5 sectors), let's see what this breaks..

7 years agodrc: move constmap out of regstat
notaz [Wed, 7 Dec 2011 16:50:19 +0000 (18:50 +0200)]
drc: move constmap out of regstat

saves close to 1MB, should help Wiz a bit

7 years agogpu_neon: fix linkage for c++ code (una-i)
notaz [Wed, 7 Dec 2011 16:34:25 +0000 (18:34 +0200)]
gpu_neon: fix linkage for c++ code (una-i)

7 years agoswitch default gpu to gpu_neon
notaz [Tue, 6 Dec 2011 23:21:27 +0000 (01:21 +0200)]
switch default gpu to gpu_neon

7 years agogpu_neon: fix several frameskip issues
notaz [Tue, 6 Dec 2011 22:51:52 +0000 (00:51 +0200)]
gpu_neon: fix several frameskip issues

7 years agogpu_neon: keep texture bits in sync
notaz [Tue, 6 Dec 2011 09:16:40 +0000 (11:16 +0200)]
gpu_neon: keep texture bits in sync

ex_regs must be updated as they are saved and sent to GPU
on frameskip

7 years agofrontend: get rid of plat_rescan_inputs
notaz [Mon, 5 Dec 2011 23:49:54 +0000 (01:49 +0200)]
frontend: get rid of plat_rescan_inputs

overannoyed by 'input: bad key' messages

7 years agoexport frame count to gpu too
notaz [Mon, 5 Dec 2011 23:19:35 +0000 (01:19 +0200)]
export frame count to gpu too

7 years agopsxcounters: avoid excessive recalculations
notaz [Mon, 5 Dec 2011 00:38:46 +0000 (02:38 +0200)]
psxcounters: avoid excessive recalculations

7 years agopsxcounters: handle overflow = target case
notaz [Mon, 5 Dec 2011 00:34:30 +0000 (02:34 +0200)]
psxcounters: handle overflow = target case

otherwise would need multiple calls to psxRcntUpdate() to get
overflow irq, but currently psxRcntUpdate() is only called once.

7 years agofrontend: show a message while BIOS is booting
notaz [Sun, 4 Dec 2011 23:06:55 +0000 (01:06 +0200)]
frontend: show a message while BIOS is booting

7 years agolet's try alternative vsync timing
notaz [Sun, 4 Dec 2011 22:27:00 +0000 (00:27 +0200)]
let's try alternative vsync timing

7 years agotry to emulate GPU busy
notaz [Sun, 4 Dec 2011 22:11:58 +0000 (00:11 +0200)]
try to emulate GPU busy

7 years agoremove gpu vblank callbacks
notaz [Sun, 4 Dec 2011 22:11:01 +0000 (00:11 +0200)]
remove gpu vblank callbacks

we do it in core now

7 years agomove some gpu status handling to core
notaz [Sun, 4 Dec 2011 21:33:50 +0000 (23:33 +0200)]
move some gpu status handling to core

this removes need to update each gpu plugin when this stuff is changed

7 years agogpu_neon: add a test
notaz [Sat, 19 Nov 2011 20:26:46 +0000 (22:26 +0200)]
gpu_neon: add a test

..that works with vram and display list dumps to produce output

7 years agogpu_neon: support unai as rendering backend
notaz [Sat, 19 Nov 2011 20:25:59 +0000 (22:25 +0200)]
gpu_neon: support unai as rendering backend

7 years agofrontend: use pld in blitters
notaz [Mon, 14 Nov 2011 21:58:15 +0000 (23:58 +0200)]
frontend: use pld in blitters

preloading a couple of cachelines ahead seems to give best results,
close to 30% speedup.

7 years agodrc/gte: decode MVMVA data deps
notaz [Tue, 1 Nov 2011 20:02:08 +0000 (22:02 +0200)]
drc/gte: decode MVMVA data deps

7 years agodrc/gte: fix dep list
notaz [Tue, 1 Nov 2011 20:00:29 +0000 (22:00 +0200)]
drc/gte: fix dep list

7 years agofrontend: use public warm path
notaz [Mon, 31 Oct 2011 13:54:45 +0000 (15:54 +0200)]
frontend: use public warm path

7 years agorelease r11 r11
notaz [Sun, 30 Oct 2011 21:42:38 +0000 (23:42 +0200)]
release r11

7 years agocdriso: support multidisk eboots
notaz [Sun, 30 Oct 2011 21:37:53 +0000 (23:37 +0200)]
cdriso: support multidisk eboots

7 years agofrontend: Wiz/Caanoo tweaks + banner image from darklight1138
notaz [Sun, 30 Oct 2011 19:31:31 +0000 (21:31 +0200)]
frontend: Wiz/Caanoo tweaks + banner image from darklight1138

7 years agofrontend: input driver and volume control for Wiz
notaz [Sun, 30 Oct 2011 19:29:01 +0000 (21:29 +0200)]
frontend: input driver and volume control for Wiz

7 years agofrontend: input: make in_ts use ts_gun
notaz [Sun, 30 Oct 2011 19:18:18 +0000 (21:18 +0200)]
frontend: input: make in_ts use ts_gun

..to stop them stealing each other's events

7 years agofrontend: input: don't configure unprobed devices
notaz [Sun, 30 Oct 2011 19:15:08 +0000 (21:15 +0200)]
frontend: input: don't configure unprobed devices

7 years agoupdate build for OABI
notaz [Sat, 29 Oct 2011 22:58:49 +0000 (01:58 +0300)]
update build for OABI

7 years agomaemo: call GPU_Open, enable bios
notaz [Sat, 29 Oct 2011 19:07:30 +0000 (22:07 +0300)]
maemo: call GPU_Open, enable bios

should fix gles breakage

7 years agofrontend: move defconfig to main
notaz [Sat, 29 Oct 2011 19:06:09 +0000 (22:06 +0300)]
frontend: move defconfig to main

.. so that maemo can use it

7 years agofix a few maemo issues
notaz [Sat, 29 Oct 2011 16:28:30 +0000 (19:28 +0300)]
fix a few maemo issues

g_opts clash, file_name, maemo_init and in_keystate

7 years agomaemo updates
Bonapart [Fri, 28 Oct 2011 16:38:46 +0000 (20:38 +0400)]
maemo updates

accelometer support for maemo
screenshot for maemo
fullscreen streghting via GTK
disabled bios(freeze emu at start)

[notasas@gmail.com] removed some unneeded stuff

7 years agofix maemo build
notaz [Sat, 29 Oct 2011 15:45:11 +0000 (18:45 +0300)]
fix maemo build

7 years agofrontend: fix screenshot functionality for pollux
notaz [Fri, 28 Oct 2011 22:15:39 +0000 (01:15 +0300)]
frontend: fix screenshot functionality for pollux

7 years agospu: add asm mixing code for ARMv5 too
notaz [Fri, 28 Oct 2011 22:09:36 +0000 (01:09 +0300)]
spu: add asm mixing code for ARMv5 too

7 years agofrontend: add touchscreen-as-buttons input code
notaz [Thu, 27 Oct 2011 23:31:57 +0000 (02:31 +0300)]
frontend: add touchscreen-as-buttons input code

intended for Caanoo/Wiz

7 years agofrontend/input: make it more modular
notaz [Thu, 27 Oct 2011 22:55:59 +0000 (01:55 +0300)]
frontend/input: make it more modular

7 years agofrontend: minor menu adjustments
notaz [Fri, 28 Oct 2011 00:04:18 +0000 (03:04 +0300)]
frontend: minor menu adjustments

7 years agofrontend: get rid of HOME hack
notaz [Thu, 27 Oct 2011 09:39:06 +0000 (12:39 +0300)]
frontend: get rid of HOME hack

nothing accesses HOME any more, not to mention this code was wrong.

7 years agogpu_neon: fix fb wrapping
notaz [Wed, 26 Oct 2011 23:18:21 +0000 (02:18 +0300)]
gpu_neon: fix fb wrapping

Tobal No.1 ...

7 years agogpu_unai: fix a few potential crashes
notaz [Wed, 26 Oct 2011 23:12:21 +0000 (02:12 +0300)]
gpu_unai: fix a few potential crashes

7 years agocdriso: add cbin support
notaz [Wed, 26 Oct 2011 22:28:11 +0000 (01:28 +0300)]
cdriso: add cbin support

7 years agofrontend/pollux: add scaling
notaz [Wed, 26 Oct 2011 21:27:28 +0000 (00:27 +0300)]
frontend/pollux: add scaling

16bpp only for now.

7 years agopsxcounters: use higher precision for vsync timing
notaz [Tue, 25 Oct 2011 23:38:58 +0000 (02:38 +0300)]
psxcounters: use higher precision for vsync timing

don't know why this is needed (having things like BIAS in mind),
maybe desyncs with cdrom code or SPU?
Tuned this on RE1.

7 years agopsxinterpreter: a bit more accurate Cause reg handling
notaz [Tue, 25 Oct 2011 21:01:47 +0000 (00:01 +0300)]
psxinterpreter: a bit more accurate Cause reg handling

7 years agodrc: detect sw irqs properly
notaz [Tue, 25 Oct 2011 21:00:05 +0000 (00:00 +0300)]
drc: detect sw irqs properly

7 years agodrc: fix various register trashing in mtc0
notaz [Tue, 25 Oct 2011 14:13:06 +0000 (17:13 +0300)]
drc: fix various register trashing in mtc0

7 years agodrc: fix cycle multiplier code
notaz [Mon, 24 Oct 2011 22:44:12 +0000 (01:44 +0300)]
drc: fix cycle multiplier code

rounding wrong way is enough to cause nasty timing problems

7 years agodrc: fix mtc0 argument
notaz [Mon, 24 Oct 2011 21:01:44 +0000 (00:01 +0300)]
drc: fix mtc0 argument

HOST_CCREG is not always cc..

7 years agodrc: some debug code improvements
notaz [Mon, 24 Oct 2011 20:13:09 +0000 (23:13 +0300)]
drc: some debug code improvements

7 years agomake root counters use generic event scheduling code
notaz [Mon, 24 Oct 2011 20:02:51 +0000 (23:02 +0300)]
make root counters use generic event scheduling code

simplifies event handling code a bit

7 years agodrc: always use physical block addresses
notaz [Sun, 23 Oct 2011 23:37:35 +0000 (02:37 +0300)]
drc: always use physical block addresses

there is no reason to use virtual addressing, things can't be remapped
on PSX as there is no MMU anyway. Fixes somewhat related invalidation
bug along the way.

7 years agorcnt: remove gate guess for cnt 0 and 1
notaz [Sun, 23 Oct 2011 20:18:19 +0000 (23:18 +0300)]
rcnt: remove gate guess for cnt 0 and 1

at least 1 is wrong according to Ryphecha's info,
and breaks at least Wipeout XL.

7 years agogte: change gte irq workaround
notaz [Sun, 23 Oct 2011 14:53:09 +0000 (17:53 +0300)]
gte: change gte irq workaround

This should be closer to what happens on the real thing.
Special thanks to Ryphecha for the idea.

7 years agodrc: detect dead gte reads too
notaz [Sun, 23 Oct 2011 14:47:38 +0000 (17:47 +0300)]
drc: detect dead gte reads too

7 years agodrc: add some hack options
notaz [Sun, 23 Oct 2011 01:17:42 +0000 (04:17 +0300)]
drc: add some hack options

intended for caanoo/Wiz, some users want to trade glitches for
performance boost.

7 years agogte: parametrize remaining ops
notaz [Fri, 21 Oct 2011 22:29:31 +0000 (01:29 +0300)]
gte: parametrize remaining ops

at least the ones from C source, it looks like there could be more.

7 years agogte: disable some MAC overflow checks for ARM
notaz [Fri, 21 Oct 2011 21:51:48 +0000 (00:51 +0300)]
gte: disable some MAC overflow checks for ARM

also add some missing ones

7 years agogte: eliminate excessive lims and casts
notaz [Fri, 21 Oct 2011 15:08:57 +0000 (18:08 +0300)]
gte: eliminate excessive lims and casts

checking for overflow where it can't occur is a waste of time,
get rid of that code.

7 years agogte: pass a pointer to registers as argument
notaz [Thu, 20 Oct 2011 23:18:46 +0000 (02:18 +0300)]
gte: pass a pointer to registers as argument

no more need to load psxRegs in each function and all registers are
reachable by ldrh/strh.

7 years agodrc: try to avoid full constant reload
notaz [Thu, 20 Oct 2011 20:58:56 +0000 (23:58 +0300)]
drc: try to avoid full constant reload

derive from existing ones if possible

7 years agodrc: fix sub assem_debug print
notaz [Thu, 20 Oct 2011 21:06:33 +0000 (00:06 +0300)]
drc: fix sub assem_debug print

7 years agodrc: avoid invcode checks on close addesses
notaz [Wed, 19 Oct 2011 23:17:22 +0000 (02:17 +0300)]
drc: avoid invcode checks on close addesses

yes this makes it possible to miss something, but I hope
it's not going to happen and will deal with it if it does.

7 years agogte_arm: some asm experiments
notaz [Wed, 19 Oct 2011 20:07:58 +0000 (23:07 +0300)]
gte_arm: some asm experiments

also removes wrong (?) MAC0 saturation.

7 years agopcnt: measure gte too
notaz [Tue, 18 Oct 2011 22:58:04 +0000 (01:58 +0300)]
pcnt: measure gte too

7 years agodrc: do MVMVA decoding; parametrize handlers
notaz [Tue, 18 Oct 2011 22:54:19 +0000 (01:54 +0300)]
drc: do MVMVA decoding; parametrize handlers

7 years agodrc: finish GTE reg liveness analysis
notaz [Sun, 16 Oct 2011 15:37:49 +0000 (18:37 +0300)]
drc: finish GTE reg liveness analysis

7 years agofrontend/pollux: program MLC2 for TV-out, save/restore regs
notaz [Fri, 14 Oct 2011 21:22:24 +0000 (00:22 +0300)]
frontend/pollux: program MLC2 for TV-out, save/restore regs

7 years agodrc: implement literal dupe merging
notaz [Wed, 12 Oct 2011 23:27:43 +0000 (02:27 +0300)]
drc: implement literal dupe merging

doesn't seem to make much difference, but theoretically should help.

7 years agoinline/parametrize rootcounter reads
notaz [Tue, 11 Oct 2011 20:03:03 +0000 (23:03 +0300)]
inline/parametrize rootcounter reads

makes rcnt1 hack impossible though, overclock PSX instead

7 years agopsxcounters: avoid doing excessive updates
notaz [Mon, 10 Oct 2011 14:38:43 +0000 (17:38 +0300)]
psxcounters: avoid doing excessive updates

this may cause missed interrupts in very rare cases,
but performance penalty this makes is surely not worth it.